draw_lift: Plot the lift curves

In thoera/metrics: Metrics for Classification Tasks

Description

Plot the lift curves.

Usage

draw_lift(data, type = c("gain_chart", "lift_curve",
  "cumulative_lift_curve"), quantile = "quantile", nb_quantiles = 20,
  lift = "lift", cumulative_lift = "cumulative_lift",
  cumulative_lift_per = "cumulative_lift_per",
  cumulative_lift_per_ideal = "cumulative_lift_per_ideal")

Arguments

data	A dataframe or a list of dataframes with the lift and the cumulative lift per quantile.
type	A string (default = "gain_chart"). The type of plot. One of: "gain_chart", "lift_curve" or "cumulative_lift_curve".
quantile	A string (default = "quantile"). The column's name of the quantiles.
nb_quantiles	An integer (default = 20). The number of quantiles computed.
lift	A string (default = "lift"). The column's name of the lift
cumulative_lift	A string (default = "cumulative_lift"). The column's name of the cumulative lift.
cumulative_lift_per	A string (default = "cumulative_lift_per"). The column's name of the cumulative lift in percentage.
cumulative_lift_per_ideal	A string (default = "cumulative_lift_per_ideal"). The column's name of the cumulative lift in percentage for the perfect model.

Value

A ggplot2 object.

See Also

compute_lift, draw_lift

Examples

data <- data.frame(
  obs = c(rep("Yes", 20), rep("No", 20)),
  prob = c(runif(n = 20, min = 0.3, max = 0.8),
           runif(n = 20, min = 0.1, max = 0.6))
)

quantiles <- qcut(data, prob = "prob")
lift <- compute_lift(quantiles)

draw_lift(lift)

thoera/metrics documentation built on Nov. 20, 2019, 2:01 p.m.